{"id":"09a81102-186b-46d6-8375-64a52be2c7a8","arxiv_id":"1908.07252","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A heated electron gas in a low-symmetry zinc-blende quantum well produces a direct current during phonon emission, an effect the authors derive and call the phonogalvanic effect.","lead":"This paper predicts that cooling a heated electron gas in a specially oriented quantum well can generate a direct electric current. The current comes from a tiny sideways shift of electrons each time they emit a phonon, which is a new mechanism for converting heat into electricity.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Equation (18) omits a (N_q+1)ℏω_q/k_B T factor, so the main result (25) is only justified in the equipartition limit.","rationale":"The paper's central claim is that energy relaxation of a heated electron gas in low-symmetry zinc-blende QWs generates a dc shift current, with Eq. (25) giving the magnitude. The most load-bearing step is the linearization in Eq. (18), because it is the bridge between the microscopic scattering rates and the final closed-form result. The exact expansion shows an omitted factor (N_q+1)ℏω_q/k_B T, which is not uniformly close to unity in the low-temperature regime relevant to HgTe QWs. This does not invalidate the existence of the effect or the qualitative sign behavior, but it does make the quantitative predictions (including the temperature dependence) conditional on the equipartition assumption. The reader's weakest assumption about phonon heating and nonthermal electron distributions is valid but broader; the Eq. (18) concern is more concrete and directly testable. Since the reader already issued a CONDITIONAL verdict and this concern reinforces conditionality rather than overturning the core argument, the verdict remains unchanged.","tokens_in":8279,"tokens_out":15874,"duration_ms":162660,"concrete_test":"Independently re-derive Eq. (18) without the equipartition approximation, keeping the exact factor (N_q+1)ℏω_q/k_B T inside the q-integration, and recompute Eq. (21) for the HgTe parameters used in Fig. 3 (d ≈ 5 nm, T = 4 K, ΔT = 10 K). If the resulting current differs from Eq. (25) by more than 30%, the quantitative prediction requires revision; if the weighted average of (N_q+1)ℏω_q/k_B T over the contributing q_z is within 10% of unity, Eq. (18) is justified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central result Eq. (25) rests on Eq. (18), which states that the emission-minus-absorption bracket equals f_{ν'}(1−f_ν) ΔT/T for ΔT ≪ T. Re-expanding the exact bracket for a transition with ε_{ν'} − ε_ν = ℏω_q gives f_ν(1−f_ν') N_q [e^{ℏω_q ΔT/(k_B T T_e)} − 1] ≈ f_ν(1−f_ν') N_q (ℏω_q/k_B T)(ΔT/T), or equivalently f_{ν'}(1−f_ν)(N_q+1)(ℏω_q/k_B T)(ΔT/T). Thus Eq. (18) is missing the factor (N_q+1)ℏω_q/k_B T, which equals unity only in the equipartition regime ℏω_q ≪ k_B T. The paper does not state this condition in Sec. III.D; the later note that the phonon energy is much smaller than the mean electron energy (below Eq. 22) is a different requirement. For HgTe/CdHgTe QWs at low temperature, q_z ∼ 1/d gives ℏω_q/k_B T of order unity, so the omitted factor is not uniformly 1. Since Eq. (18) feeds directly into Eqs. (21), (24), and (25), the predicted magnitude and temperature dependence of the shift current are affected; in particular, the absence of an explicit lattice-temperature dependence in Eq. (26) is an artifact of this approximation. The effect's existence and sign reversal between n- and p-type structures are likely robust, so this is a condition on quantitative validity rather than a fatal flaw.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes a phonogalvanic effect in low-symmetry zinc-blende quantum wells: a heated (or cooled) two-dimensional electron gas emits (or absorbs) acoustic phonons while relaxing toward the lattice temperature, and the accompanying real-space shift of Bloch electrons produces a direct electric current. The authors derive the shift for transitions between BHZ-type subband states, compute the deformation-potential scattering matrix elements, and obtain Eq. (25) for the shift current, together with the degenerate-limit expression Eq. (26). They argue that the current vanishes at equilibrium, is proportional to sin 2θ with the QW orientation angle, changes sign between n-type and p-type structures, and is of order nA/cm for ΔT = 10 K in HgTe/CdHgTe quantum wells.","tokens_in":8625,"tokens_out":5783,"duration_ms":60835,"significance":"The proposed phonogalvanic effect is conceptually interesting and connects two well-studied phenomena, energy relaxation of hot carriers and shift currents, in a way that has not, to my knowledge, been put forward before. The derivation is largely self-contained: the shift formula, the model wave functions, the deformation-potential matrix elements, and the current formula are presented in a clear sequence, and the symmetry argument for the orientation dependence sin 2θ is compelling. The explicit analytic result Eq. (25) and its sign reversal between electron and hole doping provide falsifiable predictions that could be tested in gated HgTe/CdHgTe structures. At the same time, the quantitative claim rests on an occupation-factor approximation that is not uniformly valid, as detailed below, and the numerical estimate uses an interband deformation-potential ratio borrowed from GaAs; these issues affect the magnitude but not the existence of the effect.","major_comments":[{"comment":"The small-ΔT expansion of the emission-minus-absorption bracket in Eq. (18) is incomplete. For a transition with ε_ν' − ε_ν = ℏω_q, the exact bracket is f_ν(1−f_ν') N_q [e^{ℏω_q(β−β_e)} − 1] with β = 1/k_B T and β_e = 1/k_B T_e. For ΔT = T_e − T ≪ T this expands to f_ν(1−f_ν') N_q (ℏω_q/k_B T)(ΔT/T), or equivalently f_ν'(1−f_ν) (N_q+1)(ℏω_q/k_B T)(ΔT/T). Equation (18) omits the factor (N_q+1)ℏω_q/k_B T, which equals unity only in the equipartition limit ℏω_q ≪ k_B T. This condition is not stated in Sec. III.D, and the later comment below Eq. (22) that the phonon energy is much smaller than the mean electron energy is a different requirement. Since Eq. (18) feeds directly into Eqs. (21), (24), and (25), the predicted magnitude and the lattice-temperature dependence of the shift current are affected; the apparent absence of an explicit T dependence in Eq. (26) is an artifact of this approximation.","section":"Sec. III.D, Eq. (18)"},{"comment":"The numerical estimates in Fig. 3 are obtained from Eq. (26), which is derived under the same incomplete occupation-factor expansion. For HgTe/CdHgTe quantum wells with d ~ 5 nm and low lattice temperatures, the dominant phonon wave vectors have q_z ~ 1/d, giving ℏω_q/k_B T of order unity or larger, so the equipartition condition ℏω_q ≪ k_B T is likely violated in the plotted regime. The caption specifies ΔT = 10 K but does not state the lattice temperature T. The authors should either restrict the quantitative claims to the equipartition regime, carry the corrected (N_q+1)ℏω_q/k_B T factor through the k-integration, or clearly label the numbers in Fig. 3 as illustrative estimates. The existence and sign reversal of the current are likely robust, but the central quantitative claim is not established by the present calculation.","section":"Sec. IV, Eq. (26) and Fig. 3"}],"minor_comments":[{"comment":"In the sentence after Eq. (18), 'the difference between the rates of photon emission and absorption' should read 'phonon emission and absorption'.","section":"Sec. III.D"},{"comment":"The author name 'Firdkin' should be 'Fridkin' (V. M. Fridkin).","section":"Reference [2]"},{"comment":"The phrase 'If fact' should be 'In fact'.","section":"Sec. IV, last paragraph"},{"comment":"The caption should state the lattice temperature T and explicitly note that the ratio Ξ_cv/Ξ_c is taken from GaAs rather than determined for HgTe.","section":"Fig. 3 caption"}],"recommendation":"major_revision","confidential_remarks":"The concern about Eq. (18) raised in the stress test is genuine and should be treated as a required revision. The central idea and the formal derivation are otherwise sound, but the numerical estimates should be reframed as illustrative unless the occupation-factor expansion is corrected or the equipartition condition is explicitly enforced. I see no circularity or novelty issue."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the paper identifies a real effect — a dc current generated when hot electrons relax in a low-symmetry quantum well via phonon emission, driven by the real-space shift at inelastic scattering. The symmetry reasoning and the microscopic machinery (BHZ wave functions, deformation-potential matrix elements, shift formula from Belinicher et al.) are sound, and the n- vs p-type sign reversal is a nice prediction. But the main formula as written is only justified in the equipartition limit, which the paper never states.\n\nWhat is new and what is done well: this is the first explicit calculation of a thermal (phonogalvanic) shift current in 2D zinc-blende structures, as far as I know. Prior work discussed charge shifts at inelastic scattering but did not produce a dc current from a temperature imbalance. The derivation from Eq. (5) through Eq. (25) is readable and internally consistent, and the authors correctly note that only the sum of the phase-gradient and Berry-connection contributions is gauge invariant. They also honestly flag that the interband deformation-potential ratio Ξcv/Ξc is unknown for HgTe and borrowed from GaAs.\n\nSoft spots: the stress-test concern is correct. Eq. (18) approximates the emission-minus-absorption bracket as f_{ν'}(1−f_ν) ΔT/T. The exact bracket, expanded to first order in ΔT, carries an extra factor (N_q+1)ℏω_q/k_B T (or equivalently N_qℏω_q/k_B T). That factor equals unity only when ℏω_q ≪ k_B T. The paper says phonon energies are small compared with electron energies (below Eq. 22), but that is a different condition. For the HgTe QWs considered at low T, ℏω_q/k_B T can be of order one, so the magnitude and temperature dependence in Eqs. (25)–(26) and Fig. 3 are not reliable as stated. The effect's existence and the sign reversal are robust; the quantitative curve is shakier. Also, the quasi-equilibrium single-Te assumption may need modification if phonon heating is significant, and the paper does not discuss that regime.\n\nWho this is for: people working on shift currents, phonogalvanic effects, or hot-carrier transport in non-centrosymmetric systems will find this a thought-provoking paper. It deserves a serious referee. The referee should ask for a corrected derivation of Eq. (18) with the equipartition condition stated, and for a revised numerical estimate that acknowledges the extra phonon-occupancy factor. I would not cite the numerical predictions as they stand, but I would consider citing the conceptual framework if the approximation is fixed.","headline":"A plausible new phonogalvanic shift current with a real technical flaw in the central approximation; worth refereeing, but the quantitative result needs an equipartition caveat.","tokens_in":9173,"tokens_out":5079,"would_cite":false,"duration_ms":43060,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Cooling a hot electron gas in a low-symmetry quantum well can produce a direct electric current.","keywords":["shift current","phonogalvanic effect","hot electrons","energy relaxation","Berry connection","zinc-blende quantum wells","HgTe/CdHgTe","thermoelectric current"],"falsifier":"Grow a (013)-oriented HgTe/CdHgTe quantum well, heat its electron gas with a short THz pulse while the lattice stays at base temperature, and record the DC current along the in-plane polar axis: the formula predicts a current linear in the electron-lattice temperature difference that reverses between n-type and p-type structures and disappears for [001]- and [111]-grown wells, so an experiment seeing none of these features would refute the claim.","tokens_in":8075,"feed_emoji":"⚡","tokens_out":7714,"duration_ms":74493,"temperature":0.7,"pith_summary":"The paper sets out to prove that merely cooling a hot electron gas in a semiconductor quantum well can generate a direct electric current, with no applied voltage, light, or magnetic field. It identifies a phonogalvanic effect: each phonon emission or absorption shifts the electron's real-space position, and when electrons are hotter than the lattice these shifts acquire a net direction. The central result is an analytic formula for this shift current, proportional to the electron-lattice temperature difference and to the growth-orientation factor sin 2θ, with opposite signs for electron and hole conduction. If the theory is right, a temperature imbalance alone becomes a source of DC current in low-symmetry nanostructures, with estimated nA/cm currents for modest heating and much larger currents under strong pulsed heating.","feed_headline":"Electron cooling drives a DC current in quantum wells","feed_subtitle":"Heat imbalance alone should push Bloch wave packets in low-symmetry wells into a steady current.","key_machinery":"The engine of the effect is the shift vector $\\mathbf{R}_{\\nu'\\nu} = -(\\nabla_{\\mathbf{k}'}+\\nabla_{\\mathbf{k}})\\Phi_{\\nu'\\nu} + \\boldsymbol{\\Omega}_{\\mathbf{k}'s'} - \\boldsymbol{\\Omega}_{\\mathbf{k}s}$, the real-space displacement of a Bloch electron in a transition from state $\\nu$ to $\\nu'$. It combines a phase-gradient term and a Berry-connection term, here $\\boldsymbol{\\Omega}_{\\mathbf{k}s} = s\\, b_k^2/(2k)\\, \\hat{\\mathbf{z}}\\times\\hat{\\mathbf{k}}$, and only their sum is gauge invariant. The asymmetry that makes the shifts add up comes from the electron-phonon matrix elements: an interband deformation-potential term proportional to $\\Xi_{cv}\\sin 2\\theta$ mixes the $\\Gamma_6$ and $\\Gamma_8$ bands, making phonon emission and absorption rates asymmetric in momentum and spin. This mechanism, combined with a Fermi-Dirac electron distribution at temperature $T_e$ and a Bose-Einstein phonon distribution at lattice temperature $T$, produces the current.","core_discovery":"The paper claims that energy relaxation of a heated two-dimensional electron gas in a non-centrosymmetric zinc-blende quantum well drives a direct electric current through a real-space shift of Bloch-electron wave packets at each phonon scattering event. The main result, Eq. (25), gives the shift current as\n$$j_x = \\frac{e\\hbar \\sin 2\\$\\theta$ \\, \\Xi_{cv}\\, \\$\\Delta$ T}{32\\sqrt{2}\\$pi^{2}$ \\rho $A^{3}$ $d^{3}$ T}\n\\int_\\delta^\\infty \\left[(\\Xi_c\\zeta_1+\\Xi_v\\zeta_4)\\left(2+\\frac{\\delta}{\\varepsilon_k}\\right)+\\Xi_v\\zeta_3\\left(2-\\frac{\\delta}{\\varepsilon_k}\\right)\\right]\n\\frac{\\delta}{\\varepsilon_k}\\, f_k(1-f_k)\\, d\\varepsilon_k ,$$\nwhere the integral runs over the conduction subband and an analogous expression holds for holes. The current vanishes in equilibrium, is linear in the electron-lattice temperature difference, disappears in high-symmetry [001]- and [111]-grown wells, and changes sign between n-type and p-type structures. This is a new application of the shift mechanism previously studied in photogalvanic effects, transferred to purely thermal driving.","pith_inferences":["The same shift mechanism should generalize beyond zinc-blende quantum wells to any non-centrosymmetric conductor with inelastic phonon scattering, so ferroelectric thin films or wurtzite heterostructures are natural places to look.","Cooling the electron gas below the lattice temperature should reverse the current, turning the effect into a sensitive probe of hot-carrier cooling dynamics in low-dimensional systems.","In optical experiments on the same materials, the phonogalvanic current may coexist with the photogalvanic shift current; separating them would require comparing continuous-wave illumination with pulsed-heating conditions.","Because $\\Xi_{cv}$ enters the current linearly and the paper notes its value for HgTe is not known, a measured shift current would provide a way to determine this interband deformation-potential constant."],"forward_implications":["A temperature imbalance alone acts as a source of DC current in (0lh)-oriented zinc-blende quantum wells, with no voltage, illumination, or magnetic field required.","The current reverses direction when the chemical potential moves from the electron to the hole subband, so n-type and p-type structures conduct in opposite directions.","The current is linear in $\\Delta T/T$ and in $\\sin 2\\theta$, so it vanishes in [001]- and [111]-grown wells and can be tuned by choosing the growth orientation.","For HgTe/CdHgTe parameters and $\\Delta T = 10$ K the current is of order nA/cm; under pulsed THz or optical heating, where the energy relaxation rate is orders of magnitude larger, the current should be correspondingly larger.","The current magnitude is proportional to the electron-lattice energy transfer rate, making the effect a direct electrical readout of how fast hot carriers cool."],"supporting_citations":[{"why":"Gives the real-space shift of Bloch electrons at quantum transitions in the form used for Eq. (5).","marker":"[21]"},{"why":"Provides the original derivation of the wave-packet displacement at scattering used as the starting point.","marker":"[15]"},{"why":"Supplies the k-linear two-band Hamiltonian for the electron and hole subbands in narrow-gap quantum wells.","marker":"[24]"},{"why":"Underpins the deformation-potential Hamiltonian for electron-acoustic-phonon interaction.","marker":"[27]"},{"why":"Provides the interband deformation-potential matrix elements between $\\Gamma_6$ and $\\Gamma_8$ states, the source of the asymmetry.","marker":"[29]"},{"why":"Supplies the strain-mixing matrix elements for (0lh)-oriented zinc-blende quantum wells.","marker":"[30]"},{"why":"Supplies the HgTe material parameters used in the numerical estimates.","marker":"[31]"},{"why":"Provides the k·p-model results used for band gaps and overlap integrals in HgTe/CdHgTe quantum wells.","marker":"[32]"}],"fun_headline_variants":["Heat flow alone generates DC in low-symmetry wells","Cooling electrons yields direct current in quantum wells","Thermal shift current: heat drives DC in wells","Phonon scattering turns heat into current in wells","Heat dissipation drives DC current in asymmetric wells"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The calculation assumes the hot electron gas stays internally thermalized at one effective temperature while the phonons stay at the lattice temperature, so that the emission-absorption imbalance is simply proportional to $\\Delta T/T$; if phonon heating or a nonthermal electron distribution sets in, the linear current law changes.","fun_headline_variants_meta":{"raw":{"variants":["Heat flow alone generates DC in low-symmetry wells","Cooling electrons yields direct current in quantum wells","Thermal shift current: heat drives DC in wells","Phonon scattering turns heat into current in wells","Heat dissipation drives DC current in asymmetric wells"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000507,"raw_usage":{"total_tokens":2428,"prompt_tokens":857,"completion_tokens":1571,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":473,"completion_tokens_details":{"reasoning_tokens":1497}},"tokens_in":473,"tokens_out":1571,"duration_ms":10696,"temperature":1.0,"reasoning_tokens":1497,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:22:33.334072+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Grow a (013)-oriented HgTe/CdHgTe quantum well, heat its electron gas with a short THz pulse while the lattice stays at base temperature, and record the DC current along the in-plane polar axis: the formula predicts a current linear in the electron-lattice temperature difference that reverses between n-type and p-type structures and disappears for [001]- and [111]-grown wells, so an experiment seeing none of these features would refute the claim.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the real-space shift of Bloch electrons at quantum transitions in the form used for Eq. (5)."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the original derivation of the wave-packet displacement at scattering used as the starting point."},{"cited_title":"Gantmakher and I","cited_arxiv_id":null,"evidence_quote":"Underpins the deformation-potential Hamiltonian for electron-acoustic-phonon interaction."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the interband deformation-potential matrix elements between $\\Gamma_6$ and $\\Gamma_8$ states, the source of the asymmetry."},{"cited_title":"Olbrich, C","cited_arxiv_id":null,"evidence_quote":"Supplies the strain-mixing matrix elements for (0lh)-oriented zinc-blende quantum wells."},{"cited_title":"Adachi, Handbook on Physical Properties of Semiconductors","cited_arxiv_id":null,"evidence_quote":"Supplies the HgTe material parameters used in the numerical estimates."},{"cited_title":"Dantscher, D","cited_arxiv_id":null,"evidence_quote":"Provides the k·p-model results used for band gaps and overlap integrals in HgTe/CdHgTe quantum wells."}],"review_version":1}